The long-term goal of this project is to elucidate the molecular pathogenesis underlying pediatric high- grade gliomas (HGGs). Glioblastoma is the most common and lethal type of primary brain tumor in humans, which accounts for 52% of all functional tissue brain tumor cases. Despite decades of concerted therapeutic efforts, gliomas remain incurable. The major problem is that the real pathogenesis underlying this disease is essentially unknown. Recent studies have linked pediatric and young adult HGGs with mutations in SETD2 and Gly34Arg/Val (G34R/V) substitutions in histone H3.3, but the molecular mechanism(s) by which the altered SETD2 and H3.3 drive malignancy of pediatric and young adult HGGs are not elucidated. Because SETD2 encodes the only known methyltransferase specific for histone H3 lysine36 trimethylation (H3K36me3), and H3.3G34R/V mutations lead to significant decrease in H3K36me3 levels, these observations have pointed the true culprit of pediatric and young adult HGGs to a mechanism that is regulated by H3K36me3. Strikingly, we have recently shown that H3K36me3 is essential for a critical genome-maintenance system called DNA mismatch repair (MMR) by recruiting mismatch recognition protein hMutS? to chromatin through its direct interaction with hMutS?, and that cells depleted of SETD2/H3K36me3 display a mutator phenotype usually seen in cells defective in MMR genes. We therefore hypothesize that abnormal SETD2 and H3.3 promote pediatric and young adult HGG tumorigenesis by inactivating the MMR function via blocking H3K36 trimethylation. To test this hypothesis, two specific aims are proposed in this application. Specific Aim 1 is to directly determine H3K36me3 levels and mutator phenotype in pediatric HGGs with SETD2 or H3.3G34R/V mutations. Specific Aim 2 is to measure genomic instability and the dynamic interaction between H3K36me3 and hMutS? in glioma cell lines expressing G34R/V H3.3. A successful completion of the proposed study will reveal the real pathogenesis of pediatric HGGs, providing a novel biomarker for cancer detection. More importantly, since tumor cells defective in MMR are highly resistant to many chemotherapeutic drugs including temozolomide, which is widely used for and causes resistance in glioma therapy, this study will also offer new strategies to improve glioma therapy.